How to Get Second-Order Accurate Solutions From the First-Order Accurate RELAP5 Code

Abstract

Equations of the typical ID two-fluid model of two-phase flow, which is used in today’s computer codes for the simulations in nuclear thermal-hydraulics (RELAP5, TRAC), can be written in vectorial form as:∂ψ∂t+C∂ψ∂x=P Numerical scheme in RELAP5 is based on direct discretization of the equations using first-order accurate temporal and spatial discretizations. Staggered grid is used and upwind scheme for the spatial discretization of the convection terms in the equations. Second-order accurate central differences are used only for the discretization of the pressure gradients in the momentum equations. In this paper we have demonstrated the capability of the RELAP5/MOD3 code to trace the acoustic waves with second-order accuracy, if a very small time step is chosen for the simulations. This feature of the RELAP5 code is important especially for the simulations of the fast transients describing pressure waves, i.e. shocks, rarefaction waves, water hammer... Second-order accuracy cannot be achieved for the propagation of the temperature and void fraction waves.